Harmonization of Lipoprotein(a) Immunoassays Using A Serum Panel Value Assigned with The IFCC-Endorsed Mass Spectrometry-Based Reference Measurement Procedure as A First Step Towards Apolipoprotein Standardization

Abstract

Aim: Lipoprotein (a) [Lp(a)] is a well-established risk factor for cardiovascular disease independent of low-density lipoprotein-cholesterol (LDL-C). The Lp(a) concentrations were inconsistent between the immunoassays. This study aimed to investigate whether harmonization of Lp(a) measurements can be achieved using a serum panel value assigned with the IFCC-endorsed mass spectrometry-based reference measurement procedure (IFCC-MS-RMP).

Methods: We measured the Lp(a) concentrations using five Lp(a) immunoassays in 40 panel sera provided by the Centers for Disease Control and Prevention (CDC), and 500 Japanese subjects enrolled in the Bunkyo Health Study. Of the five immunoassays, only the Roche Lp(a) assay was traceable to the WHO-IFCC reference material SRM2B. Lp(a) concentrations in CDC samples were also determined by IFCC-MS-RMP, provisionally calibrated to SRM2B. Lp(a) concentrations were expressed in mass units (mg/dL) for most reagents, but in SI units (nmol/L) for Roche’s reagent and IFCC-MS-RMP.

Results: In the CDC panel sera, all immunoassays, including Roche’s reagent, showed good correlations with IFCC-MS-RMP. In the Bunkyo Health Study samples, all immunoassays showed good correlations with Roche’s reagent (r_s, 0.986-0.998) although the slopes of the regression lines ranged from 0.292 to 0.579. After recalibration with the CDC’s panel sera, Lp(a) results of Bunkyo Health Study samples were converted to the equivalent values determined by the IFCC-MS-RMP, thus resulting in a marked reduction in the intermethod CV among the assays.

Conclusion: We achieved harmonization of Lp(a) measurements with five immunoassays using a serum panel value assigned with the IFCC-MS-RMP.

See editorial vol. 32: 563-564

1.Introduction

Numerous epidemiologic and clinical studies have consistently shown that low-density lipoprotein-cholesterol (LDL-C) is a strong risk factor for cardiovascular disease (CVD), even after adjusting for other risk factors^{1, 2)}. Although aggressive LDL-C-lowering therapy with a combination of potent statins and proprotein convertase subtilisin/kexin type 9 (PCSK9) inhibitors can significantly reduce CVD in high-risk patients, we observed persistent residual CVD risk in these patients³⁾. Among other lipoprotein-related biomarkers, many epidemiological and clinical studies have shown that lipoprotein (a) [Lp(a)] is an independent risk factor for CVD^{4, 5)} and aortic valve calcification/stenosis⁶⁾. Furthermore, the association between Lp(a) and CVD has been confirmed by recent Mendelian randomization (MR) studies using several Lp(a)-increasing single nucleotide polymorphisms (SNPs)⁷⁾. These data strongly suggest a causal role for Lp(a) in the development of CVD. Despite the impact of Lp(a) on CVD, we have been unable to achieve either harmonization or standardization owing to its molecular heterogeneity⁸⁾.

Lp(a) is a unique lipoprotein first discovered by Kare Berg in 1963 ⁹⁾. Lp(a) is a complex of LDL and apolipoprotein (a) [apo(a)], covalently linked by a disulfide bond. Apo(a) contains several tandem repeats of the plasminogen-like kringle IV (KIV) domain, followed by sequences with high homology to kringle V and protease domains of plasminogen. KIV is further classified into 10 types (designated KIV₁ to KIV₁₀). KIV₂ (type 2 of KIV) exhibits an extremely wide diversity in molecular size because the number of KIV₂ repeats ranges from 1 to ＞40, which is genetically determined by the LPA gene. Lp(a) concentrations differ across racial groups: the highest in Africans and the lowest in Asians¹⁰⁾. The majority of the population expresses two apo(a) isoforms. Currently, the Lp(a) concentrations are measured by immunoassays using antibodies against apo(a)¹¹⁾. Some assay kits use polyclonal antibodies, while others use monoclonal antibodies. It is very likely that the differences in the Lp(a) epitopes recognized by these antibodies affected the measured results.

The International Federation of Clinical Chemistry and Laboratory Medicine (IFCC) Working Group for Lp(a) Assay Standardization is establishing a reference measurement system for Lp(a) consisting of primary and secondary reference materials and a mass spectrometry (MS)-based reference measurement procedure (RMP). This RMP is endorsed by the IFCC (IFCC-MS-RMP) and expresses results in SI units (nmol/L)¹²⁾. Until reference materials are established, the IFCC-MS-RMP is calibrated using the WHO/IFCC international reference reagent material SRM2B¹³⁾. The commutability of the WHO/IFCC SRM2B has not yet been established¹¹⁾.

2.Aim

The purpose of the present study was to investigate whether we could harmonize Lp(a) measurements with different immunoassays using a serum panel value assigned with the IFCC-MS-RMP¹²⁾.

3.Methods

3. 1. Study Population

Frozen serum samples collected from two institutions were used. Five hundred serum samples were randomly collected from the participants of the Bunkyo Healthy Study, a prospective cohort study conducted over 10 years¹⁴⁾. This cohort consisted of approximately 1,000 elderly Japanese residents aged over 65 years in an urban area of Tokyo (Bunkyo-ku). Detailed background information has been described elsewhere¹⁴⁾. The second set of serum samples was obtained from volunteers with diverse ethnic backgrounds by the Centers for Disease Control and Prevention (CDC), in Atlanta, GA, the United States of America (U.S.A.). Single donor samples were collected and processed according to CLSI C37-A protocol, as described by Danilenko et al.¹⁵⁾ These unmodified serum samples had Lp(a) values assigned via the calibration laboratory in Leiden using the IFCC-MS RMP¹²⁾.

Written informed consent was obtained from the blood donors at the CDC and from participants of the Bunkyo Health Study at Juntendo University. The serum was separated by centrifugation, aliquoted into multiple polypropylene screw-capped tubes, and frozen until use. At the time of enrollment in the Bunkyo Health Study, we did not include Lp(a) in the list of measurements, because the Lp(a) assays were neither harmonized nor standardized. In 2023, we began working with the IFCC and CDC on an Lp(a) harmonization/standardization project. Therefore, we decided to use stored serum for Lp(a) measurements using the opt-out method. The Institutional Review Board of Juntendo University (initial approval no. 2015078 and latest revised version no. M15-0057-M08), we posted the protocol and contact information of this study on the website of Juntendo University so that participants of the Bunkyo Health Study could request us not to use their serum samples for this study at any time. The study protocol was reviewed and approved by each institution and conducted in accordance with the tenets of the Declaration of Helsinki¹⁶⁾.

3. 2. Lp(a) IFCC-MS RMP Method

To guarantee that commercial Lp(a) tests produce accurate results, a more recently established “IFCC Working Group on Apolipoprotein Standardization by Mass Spectrometry” developed an MS-based RMP for apo(a)¹²⁾. The RMP was operational in a network of three Calibration Laboratories. The sera used in the CDC panel were assigned by the Leiden Apolipoprotein Calibration Laboratory, Leiden, Netherlands.

3. 3. Lp(a) Immunoassays

We invited the manufacturers of Lp(a) assay kits to participate in this study through the Japan Association of Clinical Chemistry (JSCC) and the Japan Association of Clinical Reagents Industries (JACRI). Four manufacturers (Sekisui Medical, Denka, Shino-test, and Nittobo) accepted the invitation. Denka submitted two formulas: Denka-1 is the formula for the Japanese market, while Denka-2 is the formula for the foreign market. Although Denka-1 and Denka-2 use the same antibody for both kits, there are some differences in their formulas. The Shino-test submitted its prototype, which was not commercially available. We also invited Roche to participate in our study because the Lp(a) values determined using Roche’s reagent were shown to be traceable to those based on SRM2B. The Lp(a) values of the other reagents were determined using in-house standards. Of the five reagents, the Sekisui and Shino-test used monoclonal antibodies, while the others used polyclonal antibodies. The principle of the assay is turbidimetric immunoassay (TIA) for the Nittobo reagent and latex turbidimetric immunoassay (LTIA) for the others. The Roche reagent was run on Cobas pure, whereas the other reagents were run on LABOSPECT 008 (also called “Hitachi 917,” Hitachi High-Tech, Tokyo, Japan). Lp(a) results are reported in SI units (nmol/L) for the Roche reagent and in mass units (mg/dL) for the other reagents. In this study, “Lp(a)-X” was used to represent the Lp(a) value measured using Company X’s reagent. When the Lp(a) results exceeded the upper limit of the analytical range, the sample volume was automatically reduced or the samples were automatically diluted with saline. The principles and antibodies used for each kit are summarized in Supplemental Table 1.

Supplemental Table 1.Details of tested Lp(a) immunoassay kits

Reagent (IVD company)	A (Sekisui Medical)	B (Denka-1)	C (Denka-2)	D (Shino-test)	E (Nittobo Medical)	F (Roche)
Brand Name	Lp(a) Latex 「DAIICHI」	Lp(a)-LATEX “SEIKEN”	Lp(a) II-LATEX “SEIKEN”	Shino-test Lp(a) reagent (Prototype)	N-Assay TIA Lp(a) Nittobo	Tina-quant Lipoprotein (a) Gen.2
Principle	LTIA	LTIA	LTIA	LTIA	TIA	LTIA
Type of antibody	Monoclonal	Polyclonal	Polyclonal	Monoclonal	Polyclonal	Polyclonal
(Species)	(mouse)	(rabbit)#1	(rabbit) #1	(mouse)	(goat)	(rabbit)
Calibrator	Lp(a)Latex Standard Serum L,M,H	Lp(a) Calibrators#2 (Made by Denka)	Lp(a) Calibrators#2 (Made by Denka)	Lp(a) Calibrators#3 (Made by Shino-test)	TIA Lp(a) Calibrator (Multi-point)	Preciset Lp(a) Gen.2
Traceability	In-house standard	In-house standard	In-house standard	In-house standard	In-house standard	SRM 2B
(Certifciation organization)						(IFCC/WHO)
Quality control material	Lp(a) Control Serum “DAIICHI” LOW&HIGH	Lp(a) Control N Lp(a) Control AN	Lp(a) Control N Lp(a) Control AN	Lp(a) Control	Immunoquest Lp(a)-I Immunoquest Lp(a)-II	PreciControl Lp(a) Gen.2
Analyzer	LABOSPECT 008	LABOSPECT 008	LABOSPECT 008	LABOSPECT 008	LABOSPECT 008	Cobas pure c303
	(Ultrasonic mixing)	(Ultrasonic mixing)	(Ultrasonic mixing)	(Ultrasonic mixing)	(Ultrasonic mixing)	(Ultrasonic mixing)
Sample Vol. (mL)	2.0	3.2	1.8	1.7	7.0	1.5
Reagent Vol. (mL) R1/R2	120/40	120/80	120/30	100/33	140/14	100/25
Abs WL
2nd/primary (nm)	-/600	-/700	-/700	-/600	700/340	800/660
Assay Mode	2-point end assay	2-point end assay	2-point end assay	2-point end assay	2-point end assay	2-point end assay
Calibration	Spline	Spline	Spline	Spline	Line-Graph	Spline
Analytical measurement range	1-100 mg/dL	0.5-80 mg/dL	0.5-80 mg/dL	1.3-80 mg/dL	2.0-100 mg/dL	7-240 nmol/L
Automatic re-test	＞100mg/dL	＞80 mg/dL	＞80 mg/dL	＞80 mg/dL	＞100 mg/dL	＞240 mg/dL
Sample reduction	-25%	N/A	N/A	-12% (80-100 mg/dL)	-50%	N/A
Sample dilution	N/A	1:3 (saline)	1:3 (saline)	1:3 (saline, ＞100 mg/dL)	N/A	1:3 (saline)

LABOSPECT 008 is the brand name in Japan and is the same autoanalyzer sold in Europe and the U.S.A. under the brand name Hitachi 917.

IVD, in-vitro diagnotics; Vol, volume; Abs WL, absorbance wavelength

Each reagent was run on corresponding analyzers.

LTIA, latex turbidimetric immunoassay; TIA, turbidimetric immunoassay; N/A, not applicable.

#1 Denka-1 (for the Japanese market) and Denka-2 (for the foreign market) use the same antibody, although there are some differences in the formulas.

#2 Denka-1 and Denka-2 use the same calibrator made by Denka.

#3 The Shino-test reagent use their own calibrator.

Because the Roche reagent was used as a reference in the Bunkyo Health Study samples, we evaluated the limit of blank (LoB), limit of detection (LoD), and limit of quantitation (LoQ) of the Roche reagent in the preliminary experiments. In the first step, the Lp(a) value of the calibrator blank, which did not contain Lp(a), was determined 60 times. LoB was determined as the mean＋1.645× standard deviation (SD) using the non-parametric method (Supplemental Table 2). In the second step, we selected eight serum samples that were expected to have low Lp(a) near the expected LoD and measured the Lp(a) concentrations 10 times each. The composite SD was calculated as the square root of the sum of the SD² of the eight samples. The LoD was determined as LoB＋1.63×(composite SD). In the third step, we collected 13 samples whose Lp(a) concentrations were close to the expected LoQ and measured their Lp(a) concentrations in duplicate for five days. For each sample, we determined the mean, SD and coefficient of variance (CV). We plotted the CV values against Lp(a) concentration and determined 2.2 nmol/L as the LoQ at a CV value of 20%.

Supplemental Table 2.Limit of blank (LoB), limit of detection (LoD) and limit of quantitation (LoQ) of the Roche reagent

	Step 1#1	Step 2#2	Step 3
Sample	Calibrator blank
Total number of measurements	60
Mean Lp(a) (nmol/L)	0.1568
SD (nmol/L)	0.2347
LoB	0.5428
Sample (n)		Serum (n = 8)
Total number of measurements		80 (10 times each)
Mean Lp(a) for all samples (nmol/L)		2.646
Range of mean Lp(a) for each sample (nmol/L)		1.670 − 4.235
Composite SD#1 (nmol/L)		0.468
Range of SD for each sample (nmol/L)		0.2723 − 1.012
LoD (nmol/L)		1.3127
Sample (n)			Serum (n = 13)
Total number of measurements			130 (10 times each)
Mean Lp(a) for all samples (nmol/L)			6.660
Range of mean Lp(a) for each sample (nmol/L)			0.976 – 23.18
Range of SD for each sample (nmol/L)			0.272 – 1.012
Range of CV for each sample (%)			1 .3 – 74.7
LoQ, according to CV
CV 10%			4.097
CV 20%			2.220
CV 30%			1.551

Lp(a) concentration was measured in calibrator blank containing no Lp(a) (Step 1), while it was measured in serum samples containing low levels of Lp(a) (Step 2 and Step 3).

#1 The composite SD was calculated as the square root of the sum of the SD² of 8 samples. The SD value of each sample was calculated from Lp(a) values measured 10 times.

3. 4. Conversion of Lp(a) Concentrations from the Mass Units to SI Units by CDC Panel Serum

Lp(a) concentrations were measured in 40 serum panel samples collected at the CDC by participating immunoassays, including the Roche reagent. The Lp(a) concentrations in these samples were determined using the IFCC-MS-RMP with SI unit (nmol/L). Thus, the Lp(a) results are called the “Lp(a)-IFCC” and the detailed procedures of the IFCC-MS-RMP have already been published elsewhere¹²⁾.

The Lp(a) results measured by the five tested reagents were plotted against Lp(a)-IFCC. Based on the relationship between Lp(a)-X and Lp(a)-IFCC, a linear regression equation was obtained for each reagent using the least-squares method. Using the inverse functions of the individual regression equations, Lp(a) results in the mass unit were converted to those corresponding to the SI unit. These converted Lp(a) results are designated as “Lp(a)_IFCC-X.” The Lp(a) results obtained using the Roche reagent were converted in the same manner and designated as Lp(a)_IFCC-Roche.

3. 5. Assessment of Precision

The precision of each assay was evaluated using both commercially available QC controls and homemade aliquoted pooled serum samples with low and high Lp(a) concentrations. Repeatability was determined by measuring Lp(a) concentrations in each sample 20 times within one day. CV was determined using the mean and SD. The intermediate precision was determined by measuring the Lp(a) concentration in each sample twice daily for 20 consecutive days in duplicate. The mean of the two measurements per day was used to calculate the CV. The total CV was calculated as the square root of (repeatability CV)²＋(intermediate precision CV)².

3. 6. Statistical Analysis

Data are presented as the mean±SD. Because the Lp(a) concentrations were skewed to a lower value, the relationships between the different assay kits were analyzed using Spearman’s correlation coefficients. The means between the groups were compared using Student’s t-test. Statistical significance was set at p＜0.05. These calculations were performed using the add-in software program Statcel (version 4.0) in Microsoft EXCEL.

4.Results

4. 1. Basic Analytical Performance

All the tested reagents demonstrated satisfactory CV values for repeatability and intermediate precision. The total CVs was less than 3% for both the low and high Lp(a) QC controls (Table 1). In the pooled serum analysis, the total CVs ranged from 3 to 6% for the low Lp(a) serum, but were less than 3% for the high Lp(a) serum.

Table 1.Repeatability and intermediate precision, and total CVs using the quality materials and the pooled frozen sera in each assay

Reagent (Unit)	Sekisui (mg/dL)	Denka-1 (mg/dL)	Denka-2 (mg/dL)	Shino-test (mg/dL)	Nittobo (mg/dL)	Roche (nmol/L)
QC material [low Lp(a)]
Repeatability
Mean＊	13.42	24.00	23.85	20.01	21.93	42.42
SD	0.14	0.30	0.31	0.19	0.29	0.49
CV (%)	1.06	1.26	1.28	0.93	1.32	1.17
Intermediate precision
Mean＊	12.66	24.36	25.51	19.34	21.47	41.57
SD	0.23	0.56	0.59	0.22	0.30	0.58
CV (%)	1.78	2.29	2.30	1.12	1.38	1.41
Total CV (%)	2.08	2.61	2.63	1.45	1.91	1.83
QC material [high Lp(a)]
Repeatability
Mean＊	47.23	59.45	58.03	54.85	61.31	120.00
SD	0.41	0.30	0.37	0.53	0.38	0.65
CV (%)	0.86	0.50	0.64	0.96	0.62	0.54
Intermediate precision
Mean＊	46.13	57.41	57.80	52.22	60.54	116.80
SD	0.43	0.82	0.73	0.63	0.54	2.46
CV (%)	0.94	1.42	1.26	1.21	0.90	2.11
Total CV (%)	1.27	1.51	1.41	1.55	1.09	2.18
Pooled serum 1 (low Lp(a))
Repeatability
Mean＊	10.93	14.71	14.53	16.46	16.60	27.22
SD	0.15	0.22	0.25	0.17	0.27	0.45
CV (%)	1.40	1.52	1.70	1.01	1.62	1.65
Intermediate precision
Mean＊	0.96	13.96	14.50	14.81	14.83	25.59
SD	0.36	0.48	0.62	0.85	0.52	0.85
CV (%)	3.74	3.47	4.27	5.75	3.52	3.31
Total CV (%)	3.99	3.79	4.60	5.84	3.87	3.70
Pooled serum 2 (high Lp(a))
Repeatability
Mean＊	26.30	36.53	37.81	38.75	31.44	71.23
SD	0.19	0.34	0.30	0.31	0.24	0.45
CV (%)	0.73	0.93	0.80	0.80	0.77	0.63
Intermediate precision
Mean＊	23.87	35.49	38.04	36.30	31.77	69.43
SD	0.59	0.46	0.34	0.65	0.37	0.80
CV (%)	2.47	1.31	0.90	1.79	1.17	1.15
Total CV (%)	2.58	1.61	1.21	1.96	1.41	1.32

^＊The mean values of Lp(a) are presented in nmol/L for Roche reagent and in mg/dL for the other immunoassays.

The experimental protocol for the determination of the CV for repeatability and intermediate precision is described in detail in the Methods section. The total CV was calculated as the square root of (repeatability CV)² + (intermediate precision CV)².

4. 2. Correlation of Lp(a) Concentration between the Tested Reagents and IFCC-MS-RMP in CDC Samples

Lp(a) concentrations measured by the tested immunoassays, including the Roche reagent, showed a good correlation with those measured by the IFCC-MS-RMP (r_s＞0.981) (Table 2). It should be noted that Lp(a)-IFCC and Lp(a)-Roche were traceable to WHO/IFCC SRM2B, whereas the Lp(a) values of the other reagents were calibrated to each company’s in-house working calibrators. The median Lp(a) values [25^th - 75^th percentile] measured by the IFCC-MS-RMP and Roche reagent were 56.6 [19.2 - 108.5] nmol/L and 50.7 [17.1 - 100.9] nmol/L, respectively. The correlations were better in the LTIA than in the TIA (Fig.1). Although all the regression lines ran approximately through the origin, their slopes varied widely. The slope for the Denka-1 reagent was almost twice that of the Sekisui reagent. The residual plots clearly show that Lp(a) values in the low and mid-range were more accurately predicted by these linear regression models than those in the high range (Fig.2). We obtained formulas that could convert the Lp(a) values in the X reagent to those in the IFCC-MS-RMP (nmol/L).

Table 2.Parameters of the regression lines between Lp(a)-IFCC and Lp(a)-X using the CDC panel sera

Reagent (Unit)	Sekisui (mg/dL)	Denka-1 (mg/dL)	Denka-2 (mg/dL)	Shino-test (mg/dL)	Nittobo (mg/dL)	Roche (nmol/L)
rs	0.986	0.994	0.992	0.992	0.981	0.992
p-value	＜0.000001	＜0.000001	＜0.000001	＜0.000001	＜0.000001	＜0.000001
Slope	0.252	0.475	0.469	0.397	0.375	0.818
Constant	1.64	2.47	2.19	2.53	6.72	4.87
Equation#1 to convert Lp(a)-X to Lp(a)IFCC-X	= 3.77＊[C]−2.39	= 2.04＊[C]−2.77	= 2.08＊[C]−2.73	= 2.48＊[C]−5.01	= 2.40＊[C]−8.64	= 1.16＊[C]−1.94

Lp(a)-X, Lp(a) concentration measured with reagent X; Lp(a)-IFCC, Lp(a) concentration measured with IFCC-MS-RMP (IFCC-endorsed mass spectrometry-based reference measurement procedure) at Leiden University. The samples were collected from the CDC (n = 40).

Lp(a) concentrations were expressed in nmol/L (CDC and Roche) or in mg/dL (the other reagents).

Fig.1. Scatter plots of Lp(a)-X versus Lp(a)-IFCC in samples from the CDC panel sera

The Lp(a)-IFCC concentration was measured using the IFCC-MS-RMP, while the Lp(a)-X concentration was measured using individual immunoassays. Note that the Lp(a)-IFCC concentration is shown in nmol/L on the X-axis. Lp(a)-X concentration is presented in mg/dL (panels A-E) and nmol/L (F) on the Y-axis.

The regression lines and their 95% confidence intervals are presented as solid and dotted black lines, respectively. The prediction intervals are represented by solid blue lines. The blue dotted line indicates the upper limit of Lp(a) for which we did not need to reduce the sample volume or dilute it with saline.

Fig.2. Residual plots of Lp(a)-X in samples from the CDC panel sera

The Lp(a)-X concentration is shown in mg/dL (panels A-E) or nmol/L (panel F) on the X-axis. The residual was calculated by subtracting the predicted Lp(a) value from the actual Lp(a) value using a regression equation. The blue dotted line indicates the upper limit of Lp(a) for which we did not need to reduce the sample volume or dilute it with saline.

After converting the Lp(a)-X values to Lp(a)_IFCC-X values in CDC samples, we listed the samples that had a difference of more than 20 nmol/L in Lp(a)_IFCC-X among the four LTIAs. Five of the six samples had the highest Lp(a)_IFCC-X values from the first to the fifth (Table 3). For the other samples, the average Lp(a)_IFCC-X difference was 7.6±4.7 nmol/L.

Table 3.CDC panel samples whose Lp(a)_IFCC-X values differed ＞20 nmol/L by 4 LTIAs

Reagent (Unit)	Lp(a)-IFCC#1 (nmol/L)	Lp(a)IFCC-X (Converted value)
		Sekisui (nmol/L)	Denka-1 (nmol/L)	Shino-test (nmol/L)	Roche (nmol/L)	Max minus Min among 4 LTIAs
1st highest#1	246.0	175.8	204.2	242.3	175.9	70.2
2nd	234.0	227.1	233.5	212.6	207.3	26.7
3rd	194.0	223.3	220.8	170.3	227.1	56.8
4th	180.0	162.6	155.0	184.3	169.0	34.5
5th	178.0	194.7	209.5	199.5	193.4	31.5
14th	94.5	118.9	105.3	110.1	113.7	24.4
Upper Lp(a)IFCC-X for direct measurement	N/A	374.6	160.4	163.7	231.4	114.1

Lp(a)_IFCC-X, Lp(a) concentration measured with reagent X converted to the Lp(a)-IFCC equivalent value using the equations described in Table 1. Lp(a)_IFCC-X is expressed in nmol/L in all reagents.

#1 Data are presented in the highest order of Lp(a)–IFCC values.

4. 3. Correlation of the Lp(a) Concentrations between Tested Reagents and the Roche Reagent in Bunkyo Health Study Samples

We measured Lp(a) concentrations in 500 healthy elderly subjects using five immunoassays and examined the correlations of those determined by the Roche reagent with those of the other reagents. Lp(a) concentrations were skewed to lower values (Supplemental Table 3). The median Lp(a) values [25^th–75^th percentile] measured by the Roche reagent were 18.3 [7.8-39.3] nmol/L, much less than those in the CDC samples. Lp(a) concentrations measured by LTIA correlated well with those measured by the Roche reagent, with r_s ≥ 0.986 (Fig.3, Table 4, without conversion). The Lp(a) values with the Nittobo reagent correlated well with those with the Roche reagent in the middle to low concentration range but showed significant variation in the high concentration range (Fig.3, panel E). The residual plots showed that the Lp(a) values in the low and middle ranges were more accurately predicted by these linear regression models than those in the high range (Fig.4).

Supplemental Table 3.Distribution of Lp(a)-Roche concentration in Bunkyo Health Study subjects

Class interval (nmol/L)	Class center (nmol/L)	Frequency	Cumulative frequency	Relative frequency	Cumulative relative frequency
0～＜20	10	259	259	0.518	0.518
～＜40	30	104	363	0.208	0.726
～＜60	50	54	417	0.108	0.834
～＜80	70	29	446	0.058	0.892
～＜100	90	15	461	0.030	0.922
～＜120	110	8	469	0.016	0.938
～＜140	130	9	478	0.018	0.956
～＜160	150	5	483	0.010	0.966
～＜180	170	4	487	0.008	0.974
～＜200	190	3	490	0.006	0.980
～＜220	210	3	493	0.006	0.986
～＜240	230	0	493	0.000	0.986
～＜260	250	3	496	0.006	0.992
～＜280	270	3	499	0.006	0.998
～＜300	290	1	500	0.002	1.000

Lp(a) concentration was measured using the Roche’s TIA kit on a LABOSPECT 008. We used the raw data without conversion to Lp(a)_IFCC-Lp(a) concentration.

Fig.3. Scatter plots of Lp(a)-X versus Lp(a)-Roche in samples from the Bunkyo Health Study

Lp(a) concentrations were measured using latex turbidimetric immunoassays, except for the Nittobo reagent (turbidimetric immunoassay; panel E). Note that the Lp(a)-Roche concentration is shown in nmol/L on the X-axis. The Lp(a)-X concentration is shown in mg/dL on the Y-axis (panels A-E). The black and blue solid and dotted lines are the same as those in Fig. 1.

Table 4.Parameters of the regression lines between Lp(a)-Roche and Lp(a)-X using the Bunkyo Health Study samples with and without conversion to Lp(a)_IFCC-X values

Reagent	Sekisui	Denka-1	Denka-2	Shino-test	Nittobo
Lp(a)-X (mg/dL) (without conversion)
rs	0.992	0.998	0.998	0.998	0.986
p-value	＜0.000001	＜0.000001	＜0.000001	＜0.000001	＜0.000001
Slope	0.292	0.579	0.574	0.443	0.522
Constant	0.30	1.03	1.18	0.76	2.96
Lp(a)IFCC-X (nmol/L) (with conversion)
rs	0.992	0.998	-	0.994	0.986
p-value	＜0.000001	＜0.000001	＜0.000001	＜0.000001	＜0.000001
Slope	0.945	1.018	-	0.943	1.078
Constant	0.56	1.32	-	1.32	0.56

Lp(a) concentration measured with reagent X [Lp(a)-X] was converted to Lp(a)-IFCC equivalent value [Lp(a)_IFCC-X] using the equations described in Table 1.

The Lp(a)-X concentration is expressed in mg/dL for all reagents except for Roche. Lp(a)-Roche and Lp(a)_IFCC-X concentrations were expressed in nmol/L.

Fig.4. Residual plots of Lp(a)-X in samples from the Bunkyo Health Study

The Lp(a)-X concentration is shown in mg/dL on the X-axis. The residual was calculated by subtracting the predicted Lp(a) value from the actual Lp(a) value using a regression equation. The blue dotted line indicates the upper limit of Lp(a) for which we did not need to reduce the sample volume or dilute it with saline.

4. 4. Validation of the Efficacy to Convert Lp(a)-X Results to Lp(a)_IFCC-X Values using the Correlation Relationships

Using the formulas described in Table 2, we converted the Lp(a) assay results of the Bunkyo Health Study samples to values equivalent to the Lp(a) concentrations determined by the IFCC-MS-RMP. Because we did not have the data directly measured by the IFCC-MS-RMP, the Lp(a)_IFCC-X values (converted results) of the tested reagents were plotted on the y-axis against the Lp(a)_IFCC-Roche of each sample. Because the Lp(a) values were almost the same for the Denka-1 and Denka-2 reagents [Fig.1, panels B and C; Fig.3, panels B and C], we only converted the Lp(a) values of the Denka-1 reagent. After conversion, the slopes and constants of these correlation lines were almost 1.0, and zero for all reagents (Table 4).

In the Bunkyo Health Study, we had some samples with high Lp(a)_IFCC-X values that were higher than the highest Lp(a) concentration in the CDC panel serum (246 nmol/L). As expected, the Lp(a)_IFCC-X values were consistent among the tested reagents (Fig.5, panel A), especially among the LTIAs (Fig.5, panel B). The difference among LTIAs was acceptable, at least up to the 95 percentile Lp(a)_IFCC-Roche (138 nmol/L). We then calculated the inter-method CV for each sample after excluding data where the original Lp(a) concentration was below the lower limit of the analytical range or where the converted Lp(a) values were negative for any reagent. They were significantly reduced after conversion, especially in the medium and high Lp(a) ranges (Fig.6). This effect was dominant in LTIA.

Fig.5. Scatter plots of Lp(a)_IFCC-X versus Lp(a)_IFCC-Roche in samples from the Bunkyo Health Study

Lp(a) concentration was measured using five immunoassays. All Lp(a) results were converted to Lp(a)-IFCC equivalents [Lp(a)_IFCC-X] in nmol/L, using the equations described in Table 2. Data from three latex turbidimetric immunoassays (LTIA) and one turbidimetric immunoassay (TIA) are shown in Panel A, and only those from three LTIA are shown in Panel B. The dotted lines indicate the upper limits for which we did not need to reduce the sample volume or dilute it with saline.

Fig.6. Intermethod CVs before and after conversion to Lp(a)-IFCC equivalents from the Bunkyo Health Study

We used the raw Lp(a) results [Lp(a)-X] in mg/dL measured by each immunoassay (panel A) and the converted Lp(a) results [Lp(a)_IFCC-X] in nmol/L (panel B). The intermethod CV value was calculated for each sample and plotted against the mean Lp(a) values. Samples were excluded from this analysis if the original Lp(a)-X values were lower than the lower limit of any reagent or if the converted Lp(a) _IFCC-X values were negative.

4. 5. Comparison of the Lp(a) Cutoff Values in SI Units among the Tested Reagents

We converted 30 mg/dL and 50 mg/dL in each reagent (except the Roche reagent) to Lp(a)_IFCC-X values because these values are commonly used as cutoff values for diagnosing high Lp(a) concentrations in the national guidelines. Surprisingly, Lp(a) 30 mg/dL, determined by the Sekisui reagent, corresponded to approximately 50 mg/dL, as determined by other immunoassays (Table 5). To divide the subjects into low, intermediate, and high Lp(a) groups, we set tentative cutoff points at 50 and 100 nmol/L. The overall CV was calculated as the mean of the individual intermethod CVs of each group or all subjects. Before conversion, the overall CV values were similar among the three groups. After conversion, they decreased significantly in all Lp(a) groups. This was more pronounced in the medium- and high-Lp (a) groups than in the low-Lp (a) group (Table 6).

Table 5. Lp(a)_IFCC-X values that correspond to cutoff points often used as clinical decision limits

Cuoff value	Lp(a)IFCC-X values
	Sekisui	Denka-1	Denka-2	Shino-test	Nittobo
30 mg/dL	112	58	59	69	62
50 mg/dL	192	100	102	120	115

Values are expressed in nmol/L.

For each immunoassay, 30 mg/dL and 50 mg/dL were converted to Lp(a)IFCC-X values using the regression equation shown in Table 1.

Table 6. Overall CV of intermethod CVs before and after conversion to Lp(a)_IFCC-X values in the low, intermediate, and high Lp(a) groups

Group Lp(a)IFCC-Roche (nmol/L)	Low ＜50 (n = 324)	Intermediate 50 – 100 (n = 57)	High＞100 (n = 32)	All subjects (n = 413)
Overall CV (%) (LTIA + TIA)
Before conversion	30.1 (6.9)	27.5 (5.6)	24.9 (4.2)	29.2 (6.7)
After conversion	19.0 (15.3)	12.2 (7.0)	9.8 (5.3)	17.1 (14.0)
p-value	＜0.000001	＜0.000001	＜0.000001	＜0.000001
Overall CV (%) (LTIA)
Before conversion	29.1 (4.7)	28.6 (4.2)	26.6 (3.8)	28.8 (4.6)
After conversion	15.2 (13.4)	7.8 (3.8)	6.2 (3.9)	13.2 (12.3)
p-value	＜0.000001	＜0.000001	＜0.000001	＜0.000001

The subjects were divided into three groups based on Lp(a)IFCC-Roche 50 and 100 nmol/L.

The overall CV was defined as the average (SD) of the intermethod CVs of the individual samples.

LTIA, latex turbidimetric immunoassay (Sekisui, Denka-1, Shino-test); TIA, turbidimetric immunoassay (Nittobo).

5.Discussion

The present study demonstrates that we can harmonize Lp(a) measurements with different immunoassays using a serum panel value assigned with the IFCC-MS-RMP. We found that the Lp(a)-X values in both CDC panel sera (Fig.1) and Bunkyo Health Study samples (Fig.3) showed a large variation among the immunoassays. After we converted Lp(a)-X concentrations to Lp(a)-IFCC equivalents [Lp(a)_IFCC-X], the variation in Lp(a) values was significantly reduced, especially for the LTIA data (Fig.5 and 6).

Most international guidelines describe the importance of Lp(a) as a risk factor for CVD^17-20), and some guidelines recommend that Lp(a) should be measured in all adults at least once in their lifetime for accurate risk assessment and CVD stratification^{19, 20)}. They also suggested a tentative clinical decision limit (CDL) for diagnosing a high Lp(a) concentration. The guidelines use 30 mg/dL or 50 mg/dL^18-20). However, these CDLs have a potential risk of overestimation or underestimation owing to the lack of globally standardized methods for measuring Lp(a). In clinical practice, Lp(a) concentrations are measured by immunoassays using antibodies against apo(a). Some immunoassays use monoclonal antibodies, whereas others use polyclonal antibodies (Supplemental Table 1). Some experts are concerned about the accuracy of immunoassays that use polyclonal antibodies, because they may bind to KIV₂, which has a large variation in the number of repeats⁸⁾. If this is the case, it would be difficult to convert the Lp(a) concentrations determined by immunoassays to SI units. To overcome this situation, the IFCC established the first Lp(a) standardization working group in the 1990s, which developed a serum-based international reference material called IFCC SRM2B¹³⁾. The Lp(a) concentration of SRM2B was assigned by an isoform-independent enzyme-linked immunoassay (ELISA) using two monoclonal antibodies developed by Marcovina et al.²¹⁾. Although the excellent analytical performance of this ELISA has been confirmed using a large number of clinical samples²²⁾, global harmonization of Lp(a) has not yet been achieved¹¹⁾. According to the external quality assessment survey conducted in the Netherlands in 2018, the Lp(a) results still showed large interlaboratory variation. The highest Lp(a) values were almost twice as high as the lowest Lp(a) values¹¹⁾. Similar results were observed in the present study (Fig.1 and 3). Currently, SRM2B has been discontinued in Japan. The CDC announced that SRM2B was almost exhausted, and the ELISA-based KIV₂-independent method was no longer available. Consequently a “new/2^nd ” Working Group on Apolipoprotein Standardization was established in 2017 by IFCC, with the objective to meet the metrology gaps that had arisen (i.e., not only for Lp(a) but also for ApoA-I and ApoB standardization). The IFCC working group members decided to opt for SI-traceability of test results, that is, the highest order of calibration hierarchy, as explained in ISO 17511:2021, has to be developed. This was considered key for global Lp(a)/apo(a) standardization. Although other isoform-independent ELISAs using monoclonal antibodies^{23, 24)} have been reported in the literature, their suitability as candidate RMP remains unknown.

Global standardization of serum lipids has been pursued by the CDC in collaboration with the Cholesterol Reference Method Laboratory Network (CRMLN). The CRMLN consists of highly specialized calibration laboratories worldwide and was established in 1988. The IFCC working group is closely working with the CDC to standardize Lp(a) in due time using the same approaches successfully applied in the CDC/CRMLN for traditional lipids. In this study, both the CDC CRMLN network coordinator and the IFCC APO MS WG chair facilitated this Lp(a) harmonization initiative by making available the 40 RMP value-assigned serum samples to manufacturers and laboratories. The present study demonstrates that panel sera with Lp(a) values assigned by IFCC-MS-RMP work very well to reduce the difference in reactivity of antibodies to Lp(a) between immunoassays (Fig.5 and 6, Table 6). Although some researchers are skeptical about achieving harmonization of Lp(a) immunoassays that use polyclonal antibodies, our conversion method works well even with such assays, especially around the Lp(a) cutoff values for the diagnosis or management of hyper-Lp(a)-emia. It is likely that manufacturers using polyclonal antibodies assigned Lp(a) values with different multiple calibrators of their own, which were not disclosed to the users.

Recent advances in Lp(a)-lowering therapy have increased the interest in the global standardization of Lp(a) measurements. Statins, the most effective oral LDL-C-lowering agents, increase rather than decrease Lp(a) concentrations²⁵⁾. Although other lipid-lowering agents such as niacin, microsomal triglyceride transfer protein (MTP) inhibitors, anti-PCSK9 antibodies, and cholesteryl ester transfer protein (CETP) inhibitors reduce Lp(a) concentrations, the maximum reduction rate is only approximately 40%²⁶⁾. In contrast, a daily dose of 600 mg SLN360, a short interfering RNA (siRNA) targeting hepatic apo(a) synthesis, reduced Lp(a) levels by more than 90% of basal levels in the Phase 1 APOLLO trial. The Lp(a) reduction was sustained for at least 150 days²⁷⁾. Another new class of drugs, pelacarcen, an antisense oligonucleotide (ASO), achieved more than 80% Lp(a) reduction at doses of 80 and 120 mg daily²⁸⁾. The German Apheresis Registry study reported the effects of lipoprotein apheresis on CVD events in 1,430 patients with established or progressive CVD. The median LDL-C and Lp(a) levels reduced to 67.5% and 71.1%, respectively. Both major coronary events (MACE) and major noncoronary events (MANCE) were reduced by more than 70% at one year²⁹⁾. Therefore, Lp(a) harmonization/standardization should be completed before these new therapies can be used in everyday clinical practice.

We must emphasize that we should advance the global Lp(a) standardization along with the transition from the mass unit (mg/dL) to the SI unit (nmol/L) for the following reasons. First, it is metrologically correct to express the Lp(a) concentration in the SI unit because of the heterogeneity in the molecular weight of Lp(a). The molecular weight of Lp(a) was 300–800 kDa, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Similarly, the estimated molecular weight of Lp(a) from the sequence of apo(a) was 240–880 kDa³⁰⁾. Therefore, it is impossible to accurately express the Lp(a) particle concentration in mass units because of the different molecular weights of apo(a) and the variable lipid/protein composition of the particles. Second, we can implement Lp(a) values measured by different immunoassays at almost the same time during a certain pre-agreed transition period. The former success stories were the global IFCC HbA1c and IFCC enzyme standardization programs. In general, many clinicians are resistant to changes in laboratory test results, unless it is very well explained that mass results are scientifically flawed and not fit-for-clinical-purpose. Recently, the Japanese RMP for alkaline phosphatase (ALP) has changed from the JSCC method to the IFCC method within one year. The ALP (IFCC) values were almost one-third of the ALP (JSCC) values³¹⁾, but there was little confusion throughout Japan. The case of Lp(a) is much simpler than that of ALP because Lp(a) is measured by far fewer doctors than ALP in Japan. In Western Europe, routine enzyme tests in medical laboratories generate test results that are fully traceable to IFCC Reference Measurement Systems within allowable measurement uncertainty. This is a matter of leadership and education of clinicians by laboratory professionals.

We must be aware that the present study is associated with some limitations. We evaluated only 5 Lp(a) immunoassays. One was a TIA kit using a polyclonal antibody, whereas the others were four LTIA kits using monoclonal (n=2) or polyclonal (n=2) antibodies. It appears that there is less variation in LTIA kits that use monoclonal antibodies, although this is not conclusive. To overcome this weakness, we are planning a future study to examine a larger number of Lp(a) reagents available not only in Japan, but also in the world market. With such data, we may be able to suggest preliminary CDLs for the diagnosis and management of elevated Lp(a) concentrations.

6.Conclusion

In conclusion, the present study showed that the harmonization of Lp(a) measurements among different immunoassays can be achieved using a serum panel value assigned with the IFCC-MS-RMP¹²⁾. Second, laboratory professionals need to advocate proper implementation of the concept of metrological traceability of test results, especially for heterogeneous protein measurands, such as apo (a) in Lp(a). Setting a first step by harmonizing routine Lp(a) methods with the former WHO-IFCC reference material and expressing results in molar units would already harmonize test results among laboratories. Third, laboratory professionals should communicate about the relevance of unconfounded measurements and educate IVD-manufacturers, the pharmaceutical industry, and physicians on the clinical and metrology need for ultimate global standardization of Lp(a).

Disclaimer

The findings and conclusions in this report are those of the authors (s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention and the Agency for Toxic Substances and Disease Registry. The use of trade names is for identification only and does not imply endorsement by the Centers for Disease Control and Prevention, the Public Health Service, or the US Department of Health and Human Services.

Acknowledgements

This study was supported by the MEXT-Supported Program for the Strategic Research Foundation at Private Universities, 2014-2018 (S1411006) from the Ministry of Education, Culture, Sports, Science and Technology, and by the Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant No. 15K08624 to T.M. and No. 18K07477 to S.H.) from the MEXT of Japan. We would like to thank all participants and staff of the Bunkyo Health Study.

Conflict of Interest

The authors received Lp(a) immunoassay kits from Roche, Sekisui-Medical, Denka, Shino-test, and Nittobo. T.M. received honoraria for lectures from Sekisui-Medical, Denka, and Shino-tests.

Authors’ Contributions

T.M., S.H., Y.F., and A.H. designed the study, performed statistical analyses, and wrote the manuscript. S.I., M.H., and M.W. measured Lp(a) concentrations using an immunoassay. H.T., Y.T., H.W., and R.K. designed the Bunkyo Health Study, performed the blood sampling, and discussed the data. H.W.V. and C.M.C. provided advice on the study design and data analysis, and supervised the entire study. All the authors approved the final version of the manuscript.

References

• 1)  Imano H, Noda H, Kitamura A, Sato S, Kiyama M, Sankai T, Ohira T, Nakamura M, Yamagishi K, Ikeda A, Shimamoto T and Iso H: Low-density lipoprotein cholesterol and risk of coronary heart disease among Japanese men and women: The Circulatory Risk in Communities Study (CIRCS). Prev Med, 2011; 52: 381-386
• 2)  Okamura T, Kokubo Y, Watanabe M, Higashiyama A, Miyamoto Y, Yoshimasa Y and Okayama A: Low-density lipoprotein cholesterol and non-high-density lipoprotein cholesterol and the incidence of cardiovascular disease in an urban Japanese cohort study: The Suita study. Atherosclerosis, 2009; 203: 587-592
• 3)  Imran TF, Khan AA, Has P, Jacobson A, Bogin S, Khalid M, Khan A, Kim S, Erqou S, Choudhary G, Aspry K and Wu WC: Proprotein convertase subtilisn/kexin type 9 inhibitors and small interfering RNA therapy for cardiovascular risk reduction: A systematic review and meta-analysis. PLoS One, 2023; 18: e0295359
• 4)  Takahashi D, Wada H, Ogita M, Yasuda K, Nishio R, Takeuchi M, Shitara J, Tsuboi S, Dohi T, Suwa S, Miyauchi K, Daida H and Minamino T: Impact of Lipoprotein(a) as a Residual Risk Factor in Long-Term Cardiovascular Outcomes in Patients With Acute Coronary Syndrome Treated With Statins. Am J Cardiol, 2022; 168: 11-16
• 5)  Konishi H, Miyauchi K, Kasai T, Tsuboi S, Ogita M, Naito R, Sai E, Fukushima Y, Katoh Y, Okai I, Tamura H, Okazaki S and Daida H: Impact of Lipoprotein(a) as Residual Risk on Long-Term Outcomes in Patients After Percutaneous Coronary Intervention. Am J Cardiol, 2015; 115: 157-160
• 6)  Kronenberg F, Mora S, Stroes ESG, Ference BA, Arsenault BJ, Berglund L, Dweck MR, Koschinsky M, Lambert G, Mach F, McNeal CJ, Moriarty PM, Natarajan P, Nordestgaard BG, Parhofer KG, Virani SS, von Eckardstein A, Watts GF, Stock JK, Ray KK, Tokgozoglu LS and Catapano AL: Lipoprotein(a) in atherosclerotic cardiovascular disease and aortic stenosis: a European Atherosclerosis Society consensus statement. Eur Heart J, 2022; 43: 3925-3946
• 7)  Lamina C: Mendelian Randomization: Principles and its usage in Lp(a) research. Atherosclerosis, 2022; 349: 36-41
• 8)  Kronenberg F: Lipoprotein(a) measurement issues: Are we making a mountain out of a molehill? Atherosclerosis, 2022; 349: 123-135
• 9)  Berg K: A New Serum Type System in Man--the Lp System. Acta Pathol Microbiol Scand, 1963; 59: 369-382
• 10)  Pare G, Caku A, McQueen M, Anand SS, Enas E, Clarke R, Boffa MB, Koschinsky M, Wang X, Yusuf S and Investigators I: Lipoprotein(a) Levels and the Risk of Myocardial Infarction Among 7 Ethnic Groups. Circulation, 2019; 139: 1472-1482
• 11)  Ruhaak LR and Cobbaert CM: Quantifying apolipoprotein(a) in the era of proteoforms and precision medicine. Clin Chim Acta, 2020; 511: 260-268
• 12)  Ruhaak LR, Romijn F, Begcevic Brkovic I, Kuklenyik Z, Dittrich J, Ceglarek U, Hoofnagle AN, Althaus H, Angles-Cano E, Coassin S, Delatour V, Deprez L, Dikaios I, Kostner GM, Kronenberg F, Lyle A, Prinzing U, Vesper HW and Cobbaert CM: Development of an LC-MRM-MS-based cndidate reference measurement procedure for standardization of serum apolipoprotein (a) tests. Clin Chem, 2023; 69: 251-261
• 13)  Dati F, Tate JR, Marcovina SM, Steinmetz A and Stand A: First WHO/IFCC International Reference Reagent for Lipoprotein(a) for Immunoassay - Lp(a) SRM 2B. Clin Chem Lab Med, 2004; 42: 670-676
• 14)  Someya Y, Tamura Y, Kaga H, Nojiri S, Shimada K, Daida H, Ishijima M, Kaneko K, Aoki S, Miida T, Hirayama S, Konishi S, Hattori N, Motoi Y, Naito H, Kawamori R and Watada H: Skeletal muscle function and need for long-term care of urban elderly people in Japan (the Bunkyo Health Study): a prospective cohort study. BMJ Open, 2019; 9: e031584
• 15)  Danilenko U, Vesper HW, Myers GL, Clapshaw PA, Camara JE and Miller WG: An updated protocol based on CLSI document C37 for preparation of off-the-clot serum from individual units for use alone or to prepare commutable pooled serum reference materials. Clin Chem Lab Med, 2020; 58: 368-374
• 16)  World Medical A: World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA, 2013; 310: 2191-2194
• 17)  Okamura T, Tsukamoto K, Arai H, Fujioka Y, Ishigaki Y, Koba S, Ohmura H, Shoji T, Yokote K, Yoshida H, Yoshida M, Deguchi J, Dobashi K, Fujiyoshi A, Hamaguchi H, Hara M, Harada-Shiba M, Hirata T, Iida M, Ikeda Y, Ishibashi S, Kanda H, Kihara S, Kitagawa K, Kodama S, Koseki M, Maezawa Y, Masuda D, Miida T, Miyamoto Y, Nishimura R, Node K, Noguchi M, Ohishi M, Saito I, Sawada S, Sone H, Takemoto M, Wakatsuki A and Yanai H: Japan Atherosclerosis Society (JAS) Guidelines for Prevention of Atherosclerotic Cardiovascular Diseases 2022. J Atheroscler Thromb, 2024; 31: 641-853
• 18)  Li JJ, Zhao SP, Zhao D, Lu GP, Peng DQ, Liu J, Chen ZY, Guo YL, Wu NQ, Yan SK, Wang ZW and Gao RL: 2023 Chinese guideline for lipid management. Front Pharmacol, 2023; 14: 1190934
• 19)  Mach F, Baigent C, Catapano AL, Koskinas KC, Casula M, Badimon L, Chapman MJ, De Backer GG, Delgado V, Ference BA, Graham IM, Halliday A, Landmesser U, Mihaylova B, Pedersen TR, Riccardi G, Richter DJ, Sabatine MS, Taskinen MR, Tokgozoglu L, Wiklund O and Group ESCSD: 2019 ESC/EAS Guidelines for the management of dyslipidaemias: lipid modification to reduce cardiovascular risk. Eur Heart J, 2020; 41: 111-188
• 20)  Grundy SM, Stone NJ, Bailey AL, Beam C, Birtcher KK, Blumenthal RS, Braun LT, de Ferranti S, Faiella-Tommasino J, Forman DE, Goldberg R, Heidenreich PA, Hlatky MA, Jones DW, Lloyd-Jones D, Lopez-Pajares N, Ndumele CE, Orringer CE, Peralta CA, Saseen JJ, Smith SC, Jr., Sperling L, Virani SS and Yeboah J: 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation, 2019; 139: e1082-e1143
• 21)  Marcovina SM, Albers JJ, Gabel B, Koschinsky ML and Gaur VP: Effect of the number of apolipoprotein(a) kringle 4 domains on immunochemical measurements of lipoprotein(a). Clin Chem, 1995; 41: 246-255
• 22)  Marcovina SM, Albers JJ, Wijsman E, Zhang ZH, Chapman NH and Kennedy H: Differences in Lp[a] concentrations and apo[a] polymorphs between black and white Americans. J Lipid Res, 1996; 37: 2569-2585
• 23)  Marcovina SM, Navabi N, Allen S, Gonen A, Witztum JL and Tsimikas S: Development and validation of an isoform-independent monoclonal antibody-based ELISA for measurement of lipoprotein(a). J Lipid Res, 2022; 63: 100239
• 24)  Yamada S, Inoue K, Morishita R, Ogihara T, Kubono K, Kubo N, Abe A and Sakurabayashi I: A new Lp(a) assay that is unaffected by apo(a) size polymorphism. Clin Chim Acta, 1999; 287: 29-43
• 25)  Tsimikas S, Gordts P, Nora C, Yeang C and Witztum JL: Statin therapy increases lipoprotein(a) levels. Eur Heart J, 2020; 41: 2275-2284
• 26)  Boffa MB: Emerging Therapeutic Options for Lowering of Lipoprotein(a): Implications for Prevention of Cardiovascular Disease. Curr Atheroscler Rep, 2016; 18: 69
• 27)  Nissen SE, Wolski K, Balog C, Swerdlow DI, Scrimgeour AC, Rambaran C, Wilson RJ, Boyce M, Ray KK, Cho L, Watts GF, Koren M, Turner T, Stroes ES, Melgaard C and Campion GV: Single Ascending Dose Study of a Short Interfering RNA Targeting Lipoprotein(a) Production in Individuals With Elevated Plasma Lipoprotein(a) Levels. JAMA, 2022; 327: 1679-1687
• 28)  Malick WA, Goonewardena SN, Koenig W and Rosenson RS: Clinical Trial Design for Lipoprotein(a)-Lowering Therapies: JACC Focus Seminar 2/3. J Am Coll Cardiol, 2023; 81: 1633-1645
• 29)  Schettler VJJ, Neumann CL, Peter C, Zimmermann T, Julius U, Roeseler E, Heigl F, Grutzmacher P, Blume H, Vogt A and Scientific Board of GftGAWG: The German Lipoprotein Apheresis Registry (GLAR) - almost 5 years on. Clin Res Cardiol Suppl, 2017; 12: 44-49
• 30)  Furbee JW, Jr. and Fless GM: Evaluation of common electrophoretic methods in determining the molecular weight of apolipoprotein(a) polymorphs. Anal Biochem, 1996; 234: 66-73
• 31)  Furumoto K, Fujitani N, Nohara M and Hata A: Comparison of Analytical Values after Changing to the International Standardized Method for Lactate Dehydrogenase and Alkaline Phosphatase Measurements in Mouse and Rat. Vet Sci, 2022; 9: